National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

Sec. iii]   THE OASIS AND CITY OF YARKAND   87

route would help to- account for the distance recorded by Hsüan-tsang for his journey front Chia-sha or Kashgar to Chê-chü-chia. The latter territory, which the pilgrim's itinerary places at a distance of 500 li to the south-east of Kashgar, has been identified by V. de Saint-Martin, and after him by others, with Yarkand ; but Hsüan-tsang's description, as we shall see, plainly shows that Chê-chü-chia must be looked for further south, in the direction of the present Karghalik.

Though Yarkand is in all probability a place of considerable antiquity, it is difficult to trace back its name or even its existence as a town of importance previous to the Mongol conquest in the thirteenth century. In the Chinese Yüan-shih Annals relating to the period of Mongol dominion it is frequently mentioned under the names of Yeh-li-ch`ien, Ya-érh-k`an, &c., along with the cities of Kashgar and Khotan b. Marco Polo, too, on his way from Kashgar to Khotan, duly notices the ` Province of Yarcan ', where he found Nestorian and Jacobite Christians among a population generally Muhammadan, and was struck by the prevalence of goitre, still to this day a characteristic of its inhabitants °. Mirzâ Haidar supplies, perhaps, the strongest proof of the antiquity of Yarkand by the mention he makes of the excavations which Mirzâ Abâ. Bakr carried on at the site of ` the old town,' and which were believed to have yielded him great treasures'. Mirza Haidar does not specify the exact position of the area thus excavated, but is careful to note that ` it is not known whether the old town was called Yarkand or whether it had another name' 8. He adds that in the days of his ancestors ` Yarkand was a companion city to Yangi-Hisar'.

I believe that we may attribute special significance to the latter remark ; for it helps to explain why neither Hsüan-tsang, our safest guide in the old topography of this region, nor the records of the Tang period furnish any definite notice which can be referred to Yarkand. We have seen above that Hsüan-tsang's territory of Wu-sha, which lay to the east of the MurtaghAta range, extended on the south to the gin:, i.e. the Yarkand river 9. It must thus have comprised not only the greater part of the present district of Yangi-Hisar, but also the whole of the Yarkand oasis, which tradition, as well as modern administrative division, confines to the small but fertile tract on the left bank of the river after its emergence from the mountains lo.

° See Bretschneider, Med. Researches, ii. pp. 47 sq. I follow the ordinary modern pronunciation in spelling the name as Yarkand. The form Yârkand is frequent in Muhammadan texts. For an etymology see below, note 8.

° See Yule, Marco Polo, i. p. 187.

7 Comp. Târikh-i-Rashidz, pp. 256 n., 257, 296.

8 See ibid., p. 296. The name Yarkand bears a Turki look, and may contain in its first part the Turki wordyâr, which is used throughout Eastern Turkestan for the designation of ravines deeply cut into alluvial (loess) soil, and is of common occurrence in Turkestan local names (see Hedin, Reisen in Z.-A., p. 356. The meaning alluvial terrace' there given to ydr does not accord with the application of the word as observed by me from Kâshgar to Khotan). The second part is clearly the Turki word (probably of Iranian origin) pronounced hand or kant, meaning ' cultivated land, agricultural settlement ' ; comp. Shaw, Turki Vocabulary, p. 168.

9 Compare above, pp. 42 sq.

'0 This inclusion of the Yarkand oasis in Wu-sha accounts for the mention of jade made in Hsüan-tsang's above-quoted description of that territory ; see above, p. 44.

Mirzâ Haidar (Târikh-i-Rashidi, p. 298) when describing the Yarkand river's course near the town, distinctly states that jade is found in it, and this statement is confirmed by modern information. According to a notice which Mr. Macartney kindly obtained for me from Sun Ssü-yeh, Chinese Munshi of the Kashgar Agency, jade used to be dug until 50 or 6o years ago near a village, the name of which is

given as I-kan-ch`i   , and which is described

as situated to the south of Yarkand city, near or on the Yarkand river. From Muhammadju, my Yarkandi servant, I learn that jade pebbles are ` fished ' out of the river-bed in flood time near the village of Kbzumal, on the left bank of the Yarkand river, and apparently some 16 miles to the SSW. of the city. Jade mines are known to exist at several points in the mountains drained by the Zarafshân or Yarkand river, and it is thus easy to realize that jade pebbles occur also in the rubble beds deposited by the river after its entry into the plain, just as they are found in the beds and near the banks of the Yurung-kâsh and Kara-kâsh rivers of Khotan. The Hsin-chiang-chih-lio, a treatise of the eighteenth century, notes for the river passed south of Yarkand the popular designation of .Fil-ho ' jade river'; J. as., 1846, viii. p. 245.

Antiquity of Yarkand.

Yarkand during the Tang period.